STUDIES IN HISTORY Martin Lenz • Anik Waldow

AND PHILOSOPHY OF SCIENCE

Editors
VOLUME 29

General Editor:

STEPHEN GAUKROGER, University of Sydney

Contemporary Perspectives
Editorial Advisory Board:

RACHEL ANKENY, University of Adelaide

on Early Modern Philosophy
PETER ANSTEY, University ofOtago
STEVEN FRENCH, University of Leeds
KOEN VERMEIR, KathoUeke Universiteit, Leuven
OFER GAL, University of Sydney
JOHN SCHUSTER, University of New South Wales
Nature and Norms in Thought
RICHARD YEO, Griffith University

For further volumes;

www.springer.com/senes/5671
4!J Spri
ringer
64 A. LoLordo

References Chapter 5
Anstey, Peter. 201 I.John Locke and natural philosophy. Neu York: (Kford Iniversity Press.
Kant on Imagination and the Natural
Ayers. Michael. 1991. Locke, vol. 2. London: Routledge.
Bolton, Martha. 1998. Substances, substrata and names of suhsianccs in Locke's essay. In Locke.
Sources of the Conceptual
ed. Vere Chappell. New York: Oxford Universitv Pre^s.
Carson, Emily. 2005. Locke on simple and mixed modes. Locke Studies 5: 19-38.
LoLordo. Antonia. 2008. Locke's problem concerning perception. Philosopliv aiul Flienoinenological Johannes Haag
Research 72.3: 705-724.
LoLordo, Antonia. 2012. Locke's moral man. Oxford: Oxford L'nivcr^iu Press.

(t is well-known that in his theoretical philosophy Kanl puts forward an approach to

knowledge involving two 'stems' that necessarily interact to generate knowledge of
a world existing independently ofthe subject of experience. This comprises, on the
one hand, the receptivity of our sensibility, somehow responsible for intuitive repre-
sentations and, on the other hand, the spontaneity of our understanding, conceived
as a faculty of conceptual representations. It seems that the differentiation between
these two 'stems' aligns quite neatly with the distinction between natural and nor-
mative influences on our knowledge respectively.
One of the persisting questions of Kant scholarship, however, is how nature and nor-
mativity so conceived can interact in such a way as to produce empirical representations
that are simultaneously shaped by our conceptual constraints. In other words, while
conceptual resources seem to have a significant influence on those representations, at the
same time wc need to ask how this representational spontaneity can, as Wilfrid Sellars
once put it,1 be 'guided from without' by a receptive sensibility in order to guarantee that
the ensuing representations are tmly empirical. Consequently, sensibility and under-
standing, the receptive and spontaneous faculties, must interact if they are to generate
conceptual empirical representations, i.e., normativity constrained by nature.
To make this possible, Kant introduces a funher faculty in his system that, guided
by the understanding, allows subjects of experience to 'synthesize" or unify the repre-
sentational input of sensibility into conceptually shaped representations, namely, the
faculty of inuigmation. My paper will be devoted to outlining the philosophical theory
behind these ideas. Exegetically, it will focus on the Critique of Pure Reason (CPR),
while a broadly Sellarsian interpretation will serve as a systematic background. In this
way we can at least begin to do justice to the complex function ofthe imagination as
an intermediary between conceptual norms and nature in Kant's philosophy.

'Cf. Sellars (1968, 16).

J. HaagC .)
Department of Philosophy, University of Potsdam, Potsdam, Germany
e-mail: jhaag@uni-potsdam.de

M. Lenz and A. Waldow (eds.), Conlemporarv Perspectives on Early Modern Philosophy: 65

Nature and Norms in Thought. Studies in History and Philosophy of Science 29.
DOI 10.1007/978-94-007-6241-1_5, © Springer Science+Business Media Dordrecht 2013
66 J. Haag
5.1 The Faculty of Presentation
What then is the role ofthe imagination in Kant's system? In its most general form,
it is a faculty of presentation (Darstellung).2 Kant in the third Critique introduces
the concept of presentation as follows:
If the concept of an object is given, then the business of the power of judgment in using it
for cognition consists in presentation (exhibitio), i.e., in placing a corresponding intuition
beside the concept. (CPJ 5:192)
That Kant, in the context from which this quotation is taken, ascribes this activity
of presentation not to the imagination but to the power of judgment may seem rather
surprising, especially since in other places in the same text Kant is very clear that
presentation is a (central) business of imagination. For instance, in a later paragraph
ofthe Critique ofthe Power of Judgment he writes: "[T]he faculty of presentation is
the imagination" (CPJ 5:232) (another case in point can be found in the notorious
Deduction of the Analytic ofthe Aesthetic Power of Judgment, Cf. CPJ 5:287.) What
is furthermore of importance for us is Kant's emphasis on the claim that the faculty
of apprehension is "one and the same" (CPJ 5:279) with the faculty of presentation.
Apprehension, however, will in turn be clearly identified as part of the functioning
ofthe imagination. Consequently, for the purposes ofthe task at hand we are justified
in sticking to the imagination as the faculty that does the presenting and so to retum-
ing to the task of characterizing presentation.
In order to provide a Kantian answer to this question it will be helpful to explicate
the meaning of the claim that the faculty of imagination is at the same time the faculty
of apprehension. Imagination in Kant's Critical Philosophy is first and foremost the
faculty of synthesis. Its central epistemological function consists in the synthetic con-
struction of intuitions. It can do this, however, only in accordance with - and with
the help of- concepts. In a complex 'threefold synthesis' the imagination takes up
the sensibly given material into consciousness and restructures it in accordance with the
forms of intuition specific to our (human) sensibility as well as in accordance with
the categories. This complex process is called the synthesis of apprehension.*
In the course of this process the subject synthetically apprehends the sensibly
given material and out of it constructs complex representations that are then taken
as objects:
Here that which lies in the successive apprehension is considered as representation, but the
appearance that is given to me, in spite of the fact that it is nothing more than a sum of these
representations, is considered as their object, with which my concept, which I draw from
the representations of apprehension, is to agree. (A 191/B236)
Are those representations Kant's intuitions! Not quite. In the interpretation I
would like to defend they serve rather as the sensible objects to which our intuitive
2
CPJ 5:232. Cf. 28.1:235 ff.
1
At least in the 2nd edition ofthe Critique of Pure Reason. Cf. B162. In the first edition the synthesis
of apprehension is only the first part of the threefold synthesis. Cf. A98-100 and Sect. 5.3 below.
5 Kant on Imagination and the Natural Sources of the Conceptual 67
representations demonstratively refer. According to this interpretation, which in this
respect closely follows Wilfrid Sellars's elaborate interpretation of Kant's concep-
tion of intentionality, it is the basic function of intuitions to bring an object before
the mind for consideration.4 In order to fulfil this task then, intuitions presuppose as
it were sensible models construed by the imagination according to recipes of con-
struction provided by the empirical concepts in question - models that are, as Kant
writes in the Schematism, products of methods for providing concepts with pic-
tures.5 Those sensible, representational models (Sellars's image-models) - taken by
the representing subject as objects - thus turn out to be the proper candidates for
demonstrative reference. Imagination is thus a faculty that literally provides con-
cepts with pictures made from sensible material in accordance with conceptual reci-
pes for construction: it is in this sense a faculty of presentation.
Let me try to flesh out these rather sketchy remarks thus far in more detail. The
exposition will have two parts:firstly,in order to clarify how normativity shapes nature
in terms of the specific way in which the imagination synthesizes the material recep-
tively given in the presentation of objects, I will try to elucidate the concept of an
image-model and its relationship to intuitive representations in a more systematic
manner. At the same time, I will also pay closer attention to the overall picture Sellars
gives of this part of the Kantian system. In a second, more explicitly exegetical step I
will try to shed further light on the concept of synthesis presupposed in this process,
which ultimately provides us with empirical representations of objects of experience.
5.2 Image-Models
The conception outlined above regarding the relationship between image-model
and intuition relies heavily on what I take to be Sellars's late, more sophisticated6
and thoroughly Kantian conception of perception and, in particular, his concept of
an image-model that is introduced in order to make comprehensible the conceptu-
ally guided, and at the same time empirically constrained, presenting activity of
the Kantian imagination. Image-models are, in the first approximation, complex,
three-dimensional images of objects and, as such, are the result of the operation
of a conceptually guided imagination on non-conceptual sensory input.7 Now, if
4
Cf. Sellars (1978a, §48).
^140/8179/80.
6
Jay Rosenberg and John McDowell, both of whom have worked extensively on Sellars' Kant-
interpretation, share this estimation. Cf. Rosenberg (2007b, 240) and McDowell (2009, 114).
Rosenberg in this paper gives an excellent sketch of Sellars's theory of perception in general and
his conception of image-models as products of productive imagination in particular. The way he
relates the latter to Kant's account of the threefold synthesis (ibid., Fn. 12) is, however, quite prob-
lematic, as we will see below.
7
Sellars introduces this activity of the imagination by way of "phenomenological reflection" (Sellars
1978a, §§ 3,27), but the implications ofthe concepts thus gained will prove paramount and, moreover,
they are in line with what Kant has to say about the empirical activity ofthe productive imagination.
68 J. Haag
image-models were the result of a conceptual shaping of strictly non-conceptual
sensory input via the activity of what Sellars, following Kant, calls 8 the productive
imagination, then we should start by asking: what is the nature of the sensory
input fed into this shaping? Purely receptive sensations (or sense-impressions) are
conceived upon first analysis as states of perceiving subjects, which are the effect
of our senses being affected by external objects. Moreover, these states them-
selves contain the sensibilia which the perceiving subject conceives as the proper-
ties ofthe objects of perception. As sensations, they are "non-conceptual states of
consciousness... «owe of which are apperceived" (Sellars 1968, 10).9
Given this picture of the material upon which the productive imagination oper-
ates, what happens to it in the synthesizing process? The synthesizing activity ofthe
productive imagination/ormi this receptive sensory input such that it becomes the
qualitative content of a spontaneously, and hence conceptually-structured, complex
image of a three-dimensional object. This object is represented with its sensory
Properties and is pictured from the perspective of a perceiving subject. Image-
models are thus the conscious shapings of the unconscious receptive input that is
situated below the line which separates not only receptivity from spontaneity but
also sub-conscious mental states from conscious ones. 10
But in what sense is the content of those images conceptually shaped? Image-
models themselves are conceptual as well as sensory. The productive imagination
could not form image-models unless it did so according to recipes provided by the
understanding - recipes that are part of our empirical concepts of objects." Those
recipes are therefore designed to play a role exactly corresponding to Kant's con-
cept of an empirical schema: "This representation of a general procedure of imagi-
nation in providing an image of a concept, I entitle the schema of this concept."
(A 140/B179/80)
The schema is thus a conceptual recipe for forming those sense-impressions
restricted by the Kantian mathematical, though not the dynamical categories and
their corresponding transcendental schemata.12 Image-models, consequently, are the
result of a spontaneous, conceptual shaping of sensations, which in turn is the result
of our receptive faculty being affected by things-in-themselves. Consequently, the
properties of the image thus construed are only13 their sensibilia. They comprise.
8
Cf. Sellars (1968,4; 1978a, §§ 1, 28-36).
'' We will find that this statement is in need of elaboration with respect to the Kantian approach. Cf.
below Sect. 5.5.
' 0 For the metaphor of a line that is supposed to separate what is situated 'above' it (in the sphere
of the conceptual or spontaneous) and that what lies 'below' it as purely receptive cf. McDowell
(2009).
"Sellars (1978a, §31). Cf. the discussion in McDowell (2009, 114).
12
Sellars does not make this important difference explicit, but ultimately gives a description of the
role of empirical and transcendental schemata that fits this Kantian distinction when he distin-
guishes "empirical structure" from "'categorial' features" (Sellars 1978a, § 39) Cf. ibid., §§ 22,24.
For the distinction in the Critique of Pure Reason cf. AI60/B199.
''Cf. Sellars (1978a, §22).
5 Kant on Imagination and the Natural Sources ofthe Conceptual 69
however, not only the properties actually perceived, but also the merely imagined
sensible properties that we represent the object as having:
We see the cool red apple. We see it as red on the facing side, as red on the opposite side,
and as containing a volume of cool white apple flesh. We do not see of the apple its oppo-
site side, or its inside, or its internal whiteness, or its coolness, or its juiciness. But while
these features are not seen, they are not merely believed in. These features are present in
the object of perception as actualities. They are present by virtue of being imagined.
(Sellars 1978a, §21)
Image-models are a blend of features seen and features imagined, a "sensing-
cum-imaging a unified structure" (ibid., §24). They are what we take to he the
objects of which we are directly aware in sensory or, as Sellars sometimes puts it,
perceptual consciousness: "[Ajlthough the objects of which we are directly aware in
perceptual consciousness are image-models, we are not aware of them as image-
models" (ibid., § 27). The mental states, which are the takings of those image-
models as objects, are what Sellars calls 'perceptual takings'14 and what Kant - or
at least the Kant ofthe Sellarsian interpretation introduced above - calls intuitions.1''
Hence, intuitions are the takings of the image-models as objects of experience in a
demonstrative thought, and image-models in turn are, again in Kantian terminology,
the appearances that are taken to be objects in an intuition of an object of
experience.16
Intuitions are distinct from image-models since more is necessary for taking
something to be an object of experience than to merely ascribe to it sensible proper-
ties. Intuitions represent their objects with causal and dispositional properties that
we "do not see o/[those objects]... though we see them as having them" (Sellars
1978a, §22), whereas image-models only contain properties of actual or possible
sensory experience. Intuitions furthermore are representations of objects of experi-
ence whose esse essentially is not percipi, while image-models are representations
of objects whose esse is percipi in that they are essentially perspectival objects.17
Image-models are, as it were, objects without objectivity: they always incorporate
the perspective of the perceiving subject.18 The schemata that provide the recipes for
their construction are never just schemata of objects but always of objects "in such-
and-such relation to a perceiver" (Sellars 1978a, § 34).
Most importantly, intuitions are not sensory representations: they are representa-
tions that serve to make a conceptually-laden demonstrative reference to an image-
model, which is taken to be an object of experience, thereby "bringing a particular
object before the mind for its consideration" (Sellars 1978a, § 48). In an intuition
we take a complex sensory object to be an object of experience. In this way, an
intuition can serve as the subject of a perceptual judgment that guarantees direct
»Cf.e.g. Sellars(1976, §53; 1978a, § 10,50).
ls
Cf., for instance, Sellars (1976, § 24).
16
Cf. the quote above from A 191 / B 236.
17
Cf. ibid. § 28; Sellars (1976, § 51).
,8
Cf. Sellars (1978a, §28).
70 J. Haag
contact between the ensuing judgment and what we take to be (part of) the world
outside via its demonstrative aspect (though what we take to be part of the world
outside is ultimately not in any meaningful sense 'out there' l9 ).
The conceptual content of intuitions enables them to figure in Sellars's theory as
perceptual takings in the full sense of that term, namely, as having a proto-judgmental
form.20 That is why, as Sellars points out, we may think of this kind of taking as
believing, although we must think of it as 'believing in' rather than 'believing about'
or 'believing that'21:
What is taken or, if I may so put it, believed in is represented by the complex demonstrative
phrase; while that which is believed about the object is represented by the explicitly pred-
icative phrase which follows. Perceptual takings, thus construed, provide the perceiver with
perceptual subject-terms for judgments proper. (Sellars 1978a, § 10)
Again, this sharply distinguishes intuitions from image-models, which do not
serve as means of reference to objects, but as those (mis-taken22) objects them-
selves. As McDowell correctly puts it: "Sellars does not consider claim-containing
occurrences that are themselves shapings of sensory consciousness" (McDowell
2009, 122).
Naturally, there are further questions concerning the relationship between image-
models and intuitions. One may ask: can the recipe (schema) used in the generation
ofthe image-model be simply identified with the "demonstrative conceptualization"
(Rosenberg 2007a, 273) (that is, the intuition), as Jay Rosenberg claims? I am skep-
tical and would like to suggest instead that this particular use of concepts and their
correlated schemata guides the activity ofthe productive imagination, which gener-
ates both the construction of image-models and the demonstrative reference to this
image-model as an object of experience. Indeed, Sellars claims that the productive
imagination "is a unique blend of a capacity to form images in accordance with a
recipe, and a capacity to conceive of objects in a way which supplies the relevant
recipes" (Sellars 1978a, § 31).
But here the capacity to conceive of objects should not be identified with the
capacity to demonstratively refer to objects (although the latter presupposes the
former). It should be understood as the capacity to use concepts in general - a
capacity that is presupposed not only in the constmction of image-models, but
equally so in the forming of an intuitive representation of an object. This explains
why Sellars himself proceeds by speaking about concepts and schemata without
mentioning demonstrative reference:
Kant distinguished between the concept of a dog and the schema of a dog. The former
together with the concept of a perceiver capable of changing his relation to his environment
implies a family of recipes for constructing image models of perceiver-confronting-dog.
(Sellars 1978a. §31)
"'Cf. Sellars (1968, 48/9; 1963, 97).
20
Cf. Sellars (1978b, 280/1).
^'Cf. Sellars (1982, 87).
"For the conception of mis-taking cf. Sellars (1982, 109).
5 Kant on Imagination and the Natural Sources ofthe Conceptual 71
There is no reference here to the use of concepts in their capacity to intuitively
refer to an object, but only to concepts and the related schemata. Both concepts and
schemata are simply presupposed in 'providing concepts with images' and in intui-
tive reference.
I am prepared to endorse much of what Sellars has to say about the role of imagi-
nation in both the construction of image-models and the conceptualization of those
image-models in an intuition, which is in turn conceived as a complex demonstra-
tive device in the language of thought. Furthermore, I am willing to accept this
account as an adequate reading of Kant's conception of schematization in its a pos-
teriori guise. While I have defended these claims at length in other publications,23
here I will take these parts of Sellars's interpretation largely for granted - both as
philosophically enlightening and as exegetically adequate.
5.3 Synthesis
With this elucidation of the schematizing activity of the productive imagination in
mind let me now turn to the second task at hand: Kant's theory of synthesis. In the
first edition of the CPR from 1781, Kant introduces a synthesis of apprehension as
the first part of a "threefold synthesis" (A97) that not only requires apprehension,
but also a synthesis of reproduction and a synthesis of recognition. The concept of
apprehension thus helps differentiate a complex synthetic process into its different
aspects, which each have to be taken into account in the explication of any synthesis -
empirical or a priori - that can be 'brought to concepts' (cf. A78/BI03) and thus
give us conscious representations of objects. In this complex 'threefold synthesis'
the imagination takes up into consciousness the sensibly given material and restruc-
tures it in accordance both with the forms of intuitions specific to our (human)
sensibility (space and time) and in accordance with the categories of the understand-
ing (made suitable for our sensibility or, as Kant puts it, schematized). This three-
fold synthesis is thus Kant's term in the A-Deduction for the process that gives us
both image-models and intuitions in, I would argue, much the same way as outlined
above. In the second edition from 1787 this product of a synthesis brought to con-
cepts is called a combination (Verbindung):
But the concept of combination includes, besides the concept of the manifold and of its
synthesis, also the concept of the unity of the manifold. Combination is representation of
the synthetic unity ofthe manifold. (B130/1)
Combination, as a unity of synthesis according to the first edition, requires each
aspect of the threefold synthesis and, in particular, the synthesis of recognition that
unites the other syntheses by means of concepts. In the second edition, however, this
combination is achieved by the synthesis of apprehension alone (guided by apper-
ception), which is now defined as "that combination [Zusammensetzung] of the
Cf., for instance, Haag (2007, Ch. 7 and 8).
72 J. Haag
manifold in an actual empirical intuition, whereby perception, that is, empirical
consciousness ofthe intuition (as appearance), is possible" (B160).
The role ofthe concept of a synthesis of apprehension in the second edition has
therefore changed in two decisive ways. Firstly, it is broader since it now covers the
whole synthetic process leading to combination, rather than just the first part of a
synthetic process that has to be supplemented by a synthesis of reproduction and a
synthesis of recognition, as in the 'threefold synthesis' ofthe first edition. Hence, it
facilitates by itself consciousness ofthe intuition qua appearance, i.e., of a represen-
tation taken as an object rather than a representation taken as a representation.24
While, secondly, at the same time, its scope of possible application is significantly
diminished since this synthesis is now explicitly restricted to empirical synthesis.25
The second edition's synthesis of apprehension takes up impressions and synthe-
sizes them into perceptions, i.e., conscious representations.26
In what follows I will be concemed mainly with the synthesis of apprehension of
the first edition, i.e., the synthesis of apprehension as either an a priori or empirical
first step in a complex threefold synthesis, the entirety of which alone can afford us
conscious experience.271 will, accordingly, restrict my use of the term synthesis of
apprehension to Kant's use in the first edition unless otherwise indicated. Only
when speaking again at the general level of observation (with which I first began)
will I use the term "apprehension" in the broad sense required for characterizing the
imagination as a faculty of presentation and apprehension, which implies the entire
un-intentional activity of imagination-cum-understanding we are now about to
investigate in some detail.
However, even if for the purposes of this elucidation we concentrate on the nar-
rower concept of the first edition, we encounter quasi-definitional claims that do not
fit very well with the other elements ofthe threefold synthesis. The problem becomes
apparent as soon as we ask what exactly the role of the synthesis of apprehension is
within the threefold synthesis. Kant, by way of maintaining the necessity of this
aspect of the threefold synthesis, writes:
Every intuition contains in itself a manifold which can be represented as a manifold only
insofar as the mind distinguishes the time in its sequence of one impression upon another;
for each representation, in so far as it is contained in a single moment, can never be anything
but absolute unity. In order that unity of intuition may arise out of this manifold (as is
required in the representation in space) it must first be run through, and held together. This
act I name the synthesis of apprehension because it is directed immediately upon intuition,
which does indeed offer a manifold, but a manifold, which can never be brought about
[bewirkt; J.H.] as such, and as contained in a single representation, save in virtue of such a
synthesis. (A99)
24
Kant, through the addition of "as appearance", directs attention to what we have learned to call
the act/object-ambiguity.
25
This is made entirely clear in B162 Fn.
26
Cf. A320/B377.
27
What this concept is designed to cover, nevertheless, has not simply become obsolete in the sec-
ond edition. It should be taken as belonging to the 'loss' Kant refers to in the Preface to the second
edition conceming which he explicitly refers the reader back to the first edition. Cf. B XLII.
5 Kant on Imagination and the Natural Sources of the Conceptual 73
This quote contains many important clues and one inaccuracy. Firstly, the
manifold, which every intuition contains in the way outlined in the first quoted
sentence, is the manifold of an intuition that is already subject to a synthesis. Kanl
is asking for the conditions of the possibility of a given sensory consciousness of
a manifold. He is not indicating a genetic starting point for the synthesis under
investigation, in which case he would have to refer to the manifold as (synopli-
cally) presented by receptivity.
Secondly, only if those impressions are ordered in time will wc potentially distin-
guish the manifold as such. I say 'potentially' since this has to be understood as a
necessary but not a sufficient condition for the distinctive consciousness of a manifold
as a manifold. Other conditions have to be fulfilled, notably the successful synthesis
of reproduction and recognition, i.e., the two other parts of the 'threefold synthesis'.
Thirdly, if they were not ordered in this way (i.e., if there was no succession of
moments but only single moments), then every one of these moments would contain
only such representations as constitute an absolute unity. This notion of an absolute
unity refers back to the reason Kant gave for the necessity of a spontaneous synthe-
sis, namely, that otherwise no knowledge (Erkenntnis) would be possible, since in
this case "each representation were completely foreign to every other, standing apart
in isolation" (A97). It will be necessary to come back to this topic of absolute unity,
however it should be clear already that such unity cannot be the unity of a synthetic
complex.2* The only remaining alternative would thus seem to be that this unity is a
unity that is absolute because it is not complex at all, i.e., a representation that is not
itself the representation of a manifold, although it must be able to become part of
such a representation.
Fourthly, a definite complex representation is the intuition that is supposed to
unite these representations qua absolute unities into a unitary representation of this
manifold as a manifold. Although not every intuition is obviously a representation
of a complex, every intuition itself must consequently be a complex representation.
Fifthly, this unity of a complex representation is to be secured by the act of running
through (Durchlaufen) and the holding together (Zusammennehmung) of the mani-
fold. And here we run into the exegetical difficulty mentioned above: the synthesis
of apprehension in this quote is assigned two different tasks - running through and
holding together a given manifold - but only one, namely, the running through,
ultimately defines its role in the threefold synthesis.
28
Could it be, alternatively, a unity of a merely synoptic complex provided for by the purely receptive
"synopsis of sense" (A97) (cf. below Sect. 5.5)? This is also implausible because it would mean a
duplication of structures that is superfluous at best and incoherent at worst: we would have synopti-
cally structured complexes embedded as isolated parts in synthetically structured complexes. In
this case a meaningful relation between the forms of receptivity and the forms of intuition would
be lost: the hypostasis of structured receptive input would amount to rational psychology, which is
not in any way transcendentally founded. Moreover, the argument that justifies the introduction of
forms of receptivity connects them to forms of intuition and we would not have these forms of
intuition unless we had these forms of receptivity (whatever they might be). Without them, this
connection would get completely lost, which would be fatal to the overall transcendental
justification of the picture. For more on the topic of synopsis cf. below 5.5.
74 J. Haag
5.4 A 'Threefold Synthesis'
This claim about the restricted role of the synthesis of apprehension as part of the
threefold synthesis can be substantiated by investigating the further deployment of
the synthesis of apprehension in the A-Deduction and its relation to the other parts
of this complex synthetic process. As Kant makes clear in his "deduction from
below" (in A 119 ff.), this synthesis is a function of imagination, an "active faculty
for the synthesis" (A 120) that is "immediately directed upon perceptions" (ibid.).
And he continues saying: "Since imagination has to bring the manifold of intuition
into the form of an image, it must previously have taken the impressions up into its
activity, that is, have apprehended them." (ibid.)
This taking up of impressions into the activity of sponateneity is the task of the
synthesis of apprehension and is explicitly distinguished from the generation of a
"connection ofthe impressions [Zusammenhang der Eindrücke]" (Al 21): the first
operation corresponding to the running through, the second to the holding together
ofthe first definition ofthe synthesis of apprehension. For this second operation we
consequently need a further activity, this time the "reproductive faculty of imagina-
tion" (ibid.; emphasis added), that is, the synthesis of reproduction. But before
something can be reproduced, Kant seems to reason, it must be taken up into the
faculty that does the reproducing. Those two synthetic steps are indeed "inseparably
bound up with" (A102) each other so intimately that together they constitute the
"transcendental faculty of imagination" (ibid.). However, they can be abstractly dis-
tinguished as different aspects of one comprehensive synthesis.
Yet even then the synthesis remains incomplete: a further step, making the syn-
thesis truly 'threefold', is needed. This would be the synthesis of recognition. It
answers a problem on which Kant elaborates in the 'Deduction from below':
If, however, representations reproduced one another in any order, just as they happened to
come together, this would not lead to any determinate connection of them, but only to acci-
dental collocations [Haufen]; and so would not give rise to any knowledge. Their reproduction
must, therefore, conform to a rule, in accordance with which a representation connects in the
imagination with some one representation in preference to another. (A 121; emphasis JH)
To provide us with objective representations, i.e., knowledge (cognitio), the synthe-
sis must reproductively synthesize according to a rule. This rule determines how
the apprehended and reproduced representations should be 'brought into an image'
(cf. A120). According to my Sellarsian picture, the mle determines how the appre-
hended representations are united into image-models that serve as a reference for
the corresponding intuitive representation of an object of experience. This rule is there-
fore nothing else than the empirical concept (containing the schema) that guides the
synthetic activity of the productive imagination. And the aspect of the threefold synthe-
sis that ensures its being executed according to a mle is the synthesis of recognition.29
29
When introducing the synthesis of recognition Kant puts this in terms ofthe synthesis guaranteeing
the identity of a synthesized object. He phrases the solution to the problem from A 120 explicitly
in terms of the synthesis of recognition in A124. For more on the synthesis of recognition cf. Haag
(2007, 220 ff)-
5 Kant on Imagination and the Natural Sources of the Conceptual 75
This aspect makes clear why Kant can refer to the understanding in one (prominent)
place as a "faculty of rules" (A 126). However, this part ofthe threefold synthesis may
seem less inseparably bound up with the other two in the sense that it might be possible
for there to be non-rational creatures with a faculty of imagination that comprises only
the first two aspects (depending on their additional ability for associative connection),
whereas creatures capable of apprehending only in the sense under discussion would
not be able tofindtheir way around in the world.* At any rate, for finite rational beings
such as us, the synthesis has to be threefold. Otherwise we would not have intuitions of
objects of experience, i.e., objects that are synthesized in accordance with schematized
categories, as Kant aims to prove in the 'deduction from below' - the culminating part
of the A-Deduction.
These observations concerning the inseparability of the three aspects of the
threefold synthesis point to another feature of its first part: the synthesis of appre-
hension alone cannot, through simply taking up our impressions into the activity of
the imagination, generate conscious representations (perceptiones). For a represen-
tation to be conscious more is needed, as Kant forcefully contends in the argument
from associability in A121/2 where he logically ties consciousness to the possibility
of apperceptive consciousness and the related synthesis of recognition.11 Here,
Rosenberg claims that the synthesis of apprehension alone could produce image-models, while
the synthesis of recognition is only needed to afford us intuitive representations in Sellars's sense
(Rosenberg 2007b, 240). The third aspect of the threefold synthesis, the synthesis of recognition,
in his view, is reserved for "perception across time" (ibid.). (Kant indeed does talk about identity
through time in this context. But identifying this with the task of recognition amounts to confusing
a particular argument for the synthesis of recognition with the description of its contribution to the
threefold synthesis.)
Furthermore, Rosenberg's account seems misguided in a number of other ways: the synthesis
of apprehension is nothing more than the taking up into consciousness of the manifold of receptivity;
for the construction of image-models the impressions must be reproduced. Otherwise we would
have only isolated sense-impressions. (Cf. Kant's example of the drawing of a line in thought in
A102.) The synthesis of recognition likewise cannot be omitted in the generation of image-models.
It is needed for executing the recipes for construction, that are not part of the intuition of an object
of experience, say, of a dog, though they are part of the corresponding concept of dog. (This last
remark refers back to my repudiation of Rosenberg's claim that the recipe (schema) used in the
generation of the image-model should simply be identified with the "demonstrative conceptualiza-
tion" (Rosenberg 2007a, 273), that is, the intuition.)
"Cf. the letter to Marcus Herz from May 26, 1789. McDowell quotes this letter in support of his
claim that "our sensibility should be something non-rational animals also have" (McDowell 2009.
117). Accordingly, this claim is introduced by McDowell as a constraint for any successful account
of sensory consciousness.
" In brief, I take Kant to argue that representations could not be conscious unless they are associable.
For representations can be only conscious through their combination into a complex. Associability,
however, is the presupposition of such a combination. Conscious representation, hence, presupposes
associability and cannot be thought without it - which must, consequently, be conceived as a condi-
tion of the possibility of conscious representation. To this end, however, it has to be provided with
an objective, not merely subjective ground. If associability is a purely subjective ground it has to be
supplemented by a corresponding 'affinity of appearances' (A 122) that tums out to be guaranteed
by the unity of apperception, i.e., self-consciousness. Cf. Haag (2007, 241-247). The relationship
between apperception and (empirical) synthesis of recognition is discussed in ibid., 239 f.
76 J. Haag
apprehended representations become conscious only as constituent parts of an
intuition, which implies that the threefold synthesis must complete the synthesis of
complex representations (intuitions) before any consciousness of the synthesized
parts becomes possible. Only then can we, in a further abstractive step,32 become
conscious of the parts that constitute this unitary manifold as parts of a manifold.
If this consideration is correct, it will follow that the threefold synthesis itself is
merely a necessary, but not a sufficient condition of the consciousness ofthe sensi-
ble aspects of an intuition as such, even if it would of course be a necessary and
sufficient condition for sensory consciousness. Furthermore, it would put into ques-
tion the possibility of there being non-rational creatures, at least in the way brought
up above (i.e., creatures limited to the first two steps of the threefold synthesis).33
Consequently, the ascription of consciousness to non-rational animals Kant allows
for in the letter to Herz from May 1789 should be read as an ascription on a level of
abstraction different from the one at which his discussion of the threefold synthesis
is located. It is an analogous measure that presupposes conceptually synthesized
representations, and therefore cannot elucidate the conceptual means used in the
description of their synthetic constitution.34
5.5 The Synopsis of Sense
If the task ofthe synthesis of apprehension was the taking up of impressions in the
activity of imagination, why is it still called a synthesis, which originally was
introduced as an "act of putting different representations together" (A 77 / B 103)7
This may not appear so strange if we remember that the synthesis of apprehension
is merely one aspect of a complex synthetic act. The denomination thus would
reflect its being part of a synthetic act and highlight the fact that it is the imagina-
tion (the faculty of synthesis) that does the apprehending. We can perhaps shed
further light on this denomination if we turn to two closely related questions of
crucial importance for the discussion of sensory consciousness and its conceptual
shaping: (1) What are the impressions taken up into the activity of the imagination
and (2) are they changed in undergoing this procedure?
I thmk that this abstractton would have to be very much like what McDowell suggests in 2009,
119-122, What this analysis deliberately leaves out, of course, is the alternative concept ofzrecep-
me sensation. If the present account of synthesis of apprehension should turn out to be correct,
receptive sensattons obviously would not be accessible to a comparable abstractive approach.
"Thts should serve as a powerful reminder of the fact that the metaphorical talk about 'steps',
while sometimes necessary for elucidating the details ofthe threefold synthesis, is ultimately mis-
leadmg, since it suggests that each step could take place independently of the following. To call the
three parts ot this synthesis 'aspects' is more adequate
"This is meant as a clarification of my assessment of this subject matter in Haag (2007, 231 Fn).
If correct, .t wil count against the use McDowell makes of the continuity between the sensory
consciousness of animals and rational beings.
5 Kant on Imagination and the Natural Sources ofthe Conceptual 77
The first part of the question has to take into account the synopsis of sense that I
already mentioned in passing. What is this synopsis? It is a concept that appears
only in the A-Deduction, where it is sharply distinguished from any kind of synthesis.35
Synthesis is always a function ofthe imagination (i.e., a function ofthe understand-
ing taken in a certain way), whereas synopsis does not require any activity of the
imagination and hence no synthetic activity. Synopsis is a function of sense and
sense without synthesis is sense without spontaneity, in other words, sense as sheer
receptivity. Synopsis is, therefore, a function of sheer receptivity.
An exact exegetical analysis of the background of the frequently neglected36 con-
cept of synopsis is beyond the scope of this paper,37 however, I must at least sketch
some considerations relevant in this context. This is important because the synopsis
of sense proves to be the key ingredient in a conception of conceptually shaped
empirical consciousness that guarantees that the conceptual normativity of this pro-
cess is duly guided by nature 'from without'. While synopsis is not a synthesis, it
nonetheless involves some sort of structuring of given sensory material, as Kant
makes clear in the only other mention of synopsis shortly afterwards:
If each representation were completely foreign to every other, standing apart in isolation, no
such thing as knowledge would ever arise. For knowledge is a whole in which representa-
tions stand compared and connected. As sense contains a manifold in its intuition, 1 ascribe
to it a synopsis. But to such synopsis a synthesis must always correspond; receptivity can
make knowledge possible only when combined with spontaneity. (A97)
If synopsis corresponded to synthesis with respect to its being a faculty that
unites otherwise distinct and isolated representations, we could certainly conclude
that some order is already imposed on the manifold 'in intuition' by the correspond-
ing synopsis of sense.
Yet, as we learn from the Amphiboly ofthe Concepts of Reflection of the CPR,
the structuring in this case - while (like all stracturing) it has to be an instance of
determination and determinable or, as Kant puts it, form and matter - it can also
take place in an order in which form, unlike conceptual form, precedes matter. In
what sense can determination precede the determinable, i.e., form precede matter?
Kant elucidates this with admirable clarity in the Eberhard-Streitschrift, writing that
the characteristic receptivity (eigentümliche Rezeptivität) of the mind (Gemüth) is
an innate disposition to receive sensory affection.38 It is the mere disposition to react
(in no sense spontaneously but merely passively) when acted upon and affected by
things-in-themselves. This is the sense in which even the synopsis of sense can be a
priori, as Kant puts it in A94. Synopsis of sense would thus be understood, alternatively,
as the process of structuring or its result.39
,5
Cf.A94.
36
An exception is Waxman (1991, 228-225). Waxman interestingly sees as part of the role of syn-
opsis the purgation from form that I will ascribe to the Synthesis of Apprehension.
37
But compare my attempt in Haag (2007, Ch. 4).
38
Cf. 8:222.
39
The concept of synopsis shares this particular ambiguity with the concept of synthesis that can be
used (and is used by Kant) likewise to alternatively represent the process or the product of synthesis.
78 J. Haag
Although, as we have seen, in the faculty of sense form precedes matter, I now
want to turn first to the material aspect of this activity in sheer receptivity. The first
question raised by Kant's characterization of synopsis certainly concems the nature
ofthe manifold that sense receptively orders through its synopsis, i.e., its affective
input. Hardly anything more can be said about this manifold than that our receptiv-
ity is affected by it. To say more would compromise Kant's strict opposition to the
possibility of knowledge of the nature of the things-in-themselves, by which we are
affected in non-empirical affection. All we can accordingly know is that we have to
think this original manifold of affection for transcendental-philosophical reasons
related to guidedness and passivity. Thus, in cases like this, it is justified to claim
that such an affective input has to be assumed since it is a condition of the possibil-
ity of experience: without it, we would loose a necessary sense of guidance from
without. Nonetheless, we are not entitled to make any meaningful assumptions as to
the nature of this affective input.
Very similar things must be said about the products of the synopsis of sense,
which Kant calls "impressions" (A120; A121) in the 'Deduction from below', refer-
ring to the manifold of sense preceding any synthesis. We can know nothing about
the nature of these impressions, as this would mean knowing something substantial
(as opposed to purely formal or transcendental) about something that is essentially
non-synthesized. We know, however, that we do in fact have such synoptically
stmctured impressions presenting a manifold "for intuition" (BI45) that has to be
"given prior to any synthesis of understanding and independent of it" (ibid.; my
emphasis). As these quotes indicate, the purely receptive input is not completely left
out of the second edition. Earlier in its Transcendental Deduction Kant even takes
up - what in the first edition of the Critique of Pure Reason would have been - the
forms of receptivity and the products of the synopsis of sense. At the very beginning
of §15, he writes:
The manifold of representation can be given in an intuition which is purely sensible, that is,
nothing but receptivity; and the form of this intuition can lie a priori in our faculty of repre-
sentation, without being anything more than the mode in which the subject is affected.
(BI29)
Kant's picture of sensory consciousness therefore implies the existence of com-
pletely non-synthetic, non-spontaneous and a fortiori neither conceptual nor inten-
tional sensory stmctured material, namely, the synoptically structured impressions
of sheer receptivity. With Kant we can distinguish a manifold for intuition40 from an
intuition of a manifold, i.e., a manifold not for intuition (as the material on which
the synthesis of productive imagination can operate) but a conscious manifold/or us
qua subjects of experience. Notice that the products of the synopsis would be sensa-
tions completely located below the line that separates the realm of spontaneity from
that of sheer receptivity. As such, they cannot be structured by space and time as
forms of intuition in the sense elucidated above, namely, as themselves the products
of an a priori synthesis. To suggest they were would be tantamount to saying that the
^3145.
5 Kant on Imagination and the Natural Sources of the Conceptual 79
result of this supposed structuring through receptivity was a structuring through
spontaneity after all. It seems, accordingly, that the affective input would be struc-
tured by what we might along with Sellars call forms of receptivity as opposed to the
forms of intuition that form the synthetic process.41
The first part of our above question concerning the nature of the impressions
taken up into the activity ofthe imagination via the synthesis of apprehension there-
fore has a quite straightforward answer: the items taken up can only be the synopti-
cally stmctured, purely receptive sensations. If this is correct, it will imply an answer
to the second part ofthe question concerning whether those impressions are changed
in undergoing this procedure? For, what is taken up into the activity ofthe imagina-
tion by the synthesis of apprehension cannot survive this process or, to put it more
cautiously, at least cannot survive it unchanged. We settled earlier42 on a picture of
the absolute unity of the products of this taking up as being not at all complex, i.e.,
as being themselves representations that are not the representation of a manifold. It
is precisely in this spirit that Kant writes:
Apprehension by means merely of sensation occupies only an instant, if, that is. I do not
take into account the succession of different sensations. As sensation is that clement in
appearance the apprehension of which does not involve a successive synthesis .... it has no
extensive magnitude. (A167/B209)43
The receptive sensations, in being taken up into the activity of the productive
imagination, therefore lose the stmcture they had eventually gained through the
synopsis of sense, i.e., by means ofthe forms of receptivity. It is an important aspect
of the synthesis of apprehension that it purges the receptive sensations of this stmc-
ture. The resulting sensations are thus already a product of spontaneity. They are, in
this sense, spontaneous sensations. It follows that we have a second kind of sensa-
tion that must be distinguished carefully from the products of mere receptivity: the
concept of sensation as the spontaneous product of the synthesis of apprehension.
But why do these sensations have to be of absolute unity in the first place? Some
authors (including, in all likelihood, Sellars)44 have attributed this claim to Kant's
adherence to the sensualistic proclivities of his time. But this interpretation misses
the point, neglecting the interplay of form and matter (a Kantian principle of
reflection45) at this stage of the transcendental reflection: sensations can be neither
temporally nor spatially complex since they must serve as matter for the forms that
stmcture them. To be complex at this level of analysis would mean to have temporal
or spatial extension. This then is the reason for their occupying 'only an instant':
they do not have a temporal or spatial extension because otherwise they would be
41
Cf. Sellars (1968, 29). Sellars, of course, criticizes Kant for not paying attention to this very
distinction. If my interpretation is correct, however, in this context Sellars fails to do full justice to
the subtlety of Kant's approach.
42
Cf. p. 10 above.
43
Cf. A145/B184,16:662, 18:268.
44
Cf. Sellars (1968, 27) and Henrich (1976, 17).
45
Cf. A266f./B322f.
80 J. Haag
already formed. Rather, they are simple or atomistic states that contain exactly one
monadic sensory quality.
It is important to notice that the same cannot be said for the receptive sensations
as they are taken up in spontaneity. For even after they have lost their purely recep-
tive synoptic stmcture, they themselves might have properties stemming from their
being affections of our receptivity by things-in-themselves. Due to the inaccessibil-
ity of unsynthesized reality, i.e., reality-in-itself, we can neither mle out nor confirm
that they are already structured complexes, which have only gained a further layer
of stmcture from the forms of receptivity. The spontaneous sensations, on the other
hand, should be identified with the concept of sensation that Kant refers to in the
'Stufenleiter' as a conscious representation (perception) "which relates solely to the
subject as the modification of its state" (A320/B376).
This characterization is in line with Kant's first introduction of the concept of
apprehension at the very beginning ofthe A-Deduction, which classifies products of
apprehension as "modifications of the mind in intuition" (A97). And this again
seems to take up the theme from the Transcendental Aesthetics where colors, as
paradigmatic sensations, are classified as being "not properties of bodies to the intu-
ition of which they are attached, but only modifications of the sense of sight, which
is affected in a certain manner by light" (A28) and sharply distinguished from space,
which "as condition of outer objects, necessarily belongs to their appearance or
intuition" (ibid.).
And yet there are important conceptual differences: we have here, I would like to
suggest, two different ways of conceiving what - ontologically speaking - is ulti-
mately one and the same entity. The underlying ontology is the ontology of sensa-
tions as modifications of empirical subjects (whereas receptive sensations are to be
thought of as modifications of transcendental subjects). In the Transcendental
Deduction, these modifications of the empirical subject are considered from the
perspective of transcendental-philosophy, while in the quote from the Aesthetics
they are viewed from the perspective of natural science mechanistically conceived.
In transcendental perspective those modifications serve the purpose of providing
matter, which, via forms of intuition, schematized categories and empirical con-
cepts, can be formed into image-models that can then be taken (in an intuition) as
objects of experience. From the perspective of natural science they are the result of
the affection of the senses of an empirical subject brought about by the purely spa-
tio-temporally extended objects of empirical reality - an affection that can be com-
pletely described by means of mechanistic natural science (the successor concept of
matter sub specie scientia naturalis would then be "force as the concept of mechan-
ics" (Sellars 1968, 45).46)
Given the discussion of the concept of an image-model and its difference from
the concept of intuition (namely, image-models being perspectival arrangements of
sensibly given material according to a recipe, while intuitions are the taking of those
46
Sellars simply identifies this scientific concept of color as secondary quality with the transcen-
dental concept of "the real which is an object of sensation" (B 207), which, if there is something
to my considerations, amounts to a confusion.
5 Kant on Imagination and the Natural Sources ofthe Conceptual 81
very models as the objects to which we demonstratively refer), there are, strictly
speaking, two transcendental conceptions of empirical sensation that are blended in
this characterization: one as matter in appearances (i.e., as matter determined
through forms of intuition) and another as matter for a represented object. The first
would come into play in the construction of image-models, the second in the intui-
tive reference to objects of experience.
Be that as it may, the fact remains that any class of sensations (given it has a
certain manifold) could fulfill the role of matter in this context: unlike the forms of
intuition (space and time), the concrete character of sensation is completely arbi-
trary. From the transcendental perspective we only need some matter - whatever
quality it might encompass.47 Those modifications can therefore play their transcen-
dental role while being nothing but largely arbitrary modifications of empirical sub-
jects. We can conceive of them either (taking a transcendental viewpoint) as a result
of the synthetic taking-up of the stmctured sensations of receptivity by the transcen-
dental subject or (in describing empirical reality as reality) as produced by the
empirical subject's being affected by objects that are located in space and time (just
like ourselves qua empirical subjects).
In both perspectives, however, the sensations would, ontologically speaking, be
modifications of the empirical subject. For even from the transcendental viewpoint
they are conceived as modifications of the subject taken as an empirical subject
(since they would be conscious representations and there can be no consciousness
of the modifications of the transcendental subject). As modifications of a subject,
conscious representations must be appearances, whereas the existence of the con-
scious subject is not mere appearance.48
Where does that leave us concerning the continuity between the sensations of
receptivity and the sensations of spontaneity? Talk of the synthesized sensations
being the same sensations as the receptive sensations tums out to be profoundly
inadequate for a number of reasons. Firstly, we cannot know anything about the
existence of a meaningful continuity between the two kinds of sensation due to the
inaccessibility of the transcendental subject's modifications as they are in themselves.
Secondly, there need not be any continuity conceming their qualitative content
between receptive sensations and the products of the synthesis of apprehension,
which are the result of the completely arbitrary subjective disposition of an empiri-
cal subject. And, thirdly, since sensations that are products of this kind of synthesis
are not in any way formally structured, continuity in form is likewise excluded.
Continuity in form being excluded, qualitative continuity remains completely
arbitrary and ultimately unnecessary. Moreover, given the items that could answer
this question are inaccessible, there seems to be no meaningful way left of talking
about a continuity between both kinds of sensations: we cannot say that spontane-
ous sensations are sensations of receptivity taken as something else. Whatever the
47
Sellars, of course, would deny just that, since he sees space and color as inextricably bound up
with each other. But Kant has an interesting argument against this to the effect that we only need
some matter, but exactly this kind of form. For a discussion compare Haag (2007, 142-150).
48
Cf.B 157/8.
82 J. Haag
material basis for the synthesis of apprehension, it is changed so radically that there
is no way back from synthesized sensation to receptive sensation - not even by
(Sellarsian) analogy.
5.6 Synthesis a Priori and the Concept of Guidance
But is this separation not too strict? Through the synthetic process that leads to the
construction of image-models and, in the last instance, objects of experience, we
certainly have provided for the normative-conceptual structuring of the world-as-
perceived. But, in so doing, are we not losing all philosophically interesting con-
tact with the happenings below the line that separates spontaneity and receptivity
in this way? Will not, in other words, the loss of a meaningful way of talking about
the same sensations be linked with the loss of any meaningful concept of guidance
from without?
To find a way to prevent this, we must invoke the a priori dimension of the syn-
thesis of apprehension and, with it, the formal aspects above and below this line. In
the first edition's Transcendental Deduction Kant is very explicit about the a priori
function of the synthesis of apprehension. Without such a synthesis, he writes, "we
should never have a priori the representations either of space or time" (A99). And
he continues: "They can be produced only through the synthesis of the manifold
which sensibility presents in its original receptivity. We have thus a pure synthesis
of apprehension." (A99/100)
Without such a synthesis we would have no relation to what is presented through
'original receptivity' to the subject of experience and hence no consciousness of
either space or time. In the words of the B-Deduction: we would have "forms of
intuition" (B160 fn.), but no "formal intuition" (ibid.) of space and time. I interpret
Kant's distinction between the forms of intuition and formal intuition as corre-
sponding to my Sellarsian distinction between forms of receptivity and forms of
intuition respectively.49
The synthesis of apprehension, accordingly, must have an a priori use, since our
forms of intuition are a priori forms. In this pure synthesis a manifold is somehow
synthesized into representations of space and time.50 Let me now give a brief sketch
of this process of a priori synthesis since the way it ought to be conceived will serve
to provide a way out of the problem outlined above of our losing contact with the
'below the line' elements of sensory consciousness.
First of all, it must be acknowledged that the synthetic process cannot provide us
with intuitive representations of space and time unless we expand it (in the spirit of
""Since Kant often uses form of intuition for what he calls formal intuition in B160 fn., I have
decided not to use his somewhat confusing terminology in this case, but rather to stick to Sellars's
terminological suggestion instead.
50
These (undetermined) intuitive representations of space and time have to be distinguished care-
fully from the (determined) concepts of space and time based upon them. Cf. e.g. A25/B39.
5 Kant on Imagination and the Natural Sources of the Conceptual 83
the B-Deduction) to encompass the entirety of the threefold synthesis - or, at the
very least, apprehension and reproduction must be taken together.51 Furthermore,
we would also seem to need the synthesis of recognition, as the synthesized repre-
sentations would then be conscious representations. This is, however, not a major
problem since we have to think the aspects of the threefold synthesis together
anyway.
The problem with an a priori synthesis of a manifold is that it is not easy to see
on what kind of matter this synthesis should operate. We do not have any formal
manifold in the sense of a given manifold of forms.52 The forms of receptivity are
not given, save by their product, the receptive sensations. The only manifold that
provides material for a synthesis of apprehension is therefore the formed material
that is given to our spontaneity by the synopsis of sense - a thoroughly a posteriori
product. How can we get from there to a pure synthesis of an a priori manifold? The
solution seems to lie in a use of abstraction that, as it were, purifies the synoptically
given manifold by concentrating on the process of synthesis and neglecting its mate-
rial content. Kant illustrates this concentration on the manner or mode in which the
material is given in apprehension in § 24 of the B-Deduction:
We cannot represent the three dimensions of space save by setting three lines at right angles
to one another from the same point. Even time itself we cannot represent, save in so far as
we attend, in the drawing of a straight line (which has to serve as the outer figurative repre-
sentation of time), merely to the act of the synthesis of the manifold whereby we succes-
sively determine inner sense, and in so doing attend to the succession of this determination
in inner sense. Motion, as an act of a subject (not as a determination of an object), and
therefore the synthesis ofthe manifold in space, first produces the concept of succession - if
we abstract from this manifold and attend solely to the act through which we determine the
inner sense according to its form. (B154)
What Kant describes here is not an abstract separation of the formal and qualita-
tive content of a given representation, but rather a concentration on the procedural
character of the act of synthesis. That he talks about the concept of succession and
not about time as a form of intuition (formal intuition) should not deter us from the
application of the method described to the intuition of time (or space), since this
undetermined intuitive representation, even on Kant's description, would be a nec-
essary precondition of the determined concept of time as succession.
In the act of synthesis we therefore have to concentrate on the properties that
belong solely to the formal aspects of sensibility. To this end, we abstract from the
qualitative aspects apprehended in the synthesis as well as from those formal aspects
that do not belong to sensibility but solely to understanding (and which are, conse-
quently, not aspects of the act that gives us forms of intuition but rather a priori
concepts). If we proceed in this way, we are left with what amounts to formal
intuitions of time and space. This abstractive account is therefore able to explain
51
Which could, by the way, help to explain the lapse on Kant's part in the original characterization
of the synthesis of apprehension: he might have been preoccupied with a priori synthesis.
52
Vaihinger tried to defend this line of interpretation. R.R Wolff interpreted Kant along similar
lines, but correctly judged it a failure - unfortunately, however, as a failure on Kant's part, and not
on the part of his interpretation. Cf. Vaihinger (1892, 102-111) and Wolff (1963. 218-223).
84 J. Haag
why we come to have pure intuitive representations of space and time only after
using them as forms of intuition in the schematizing of the categories in the model-
ing of image-models. These intuitions are a priori since we understand through the
act of abstraction that they are the forms that underlie each and every intuitive rep-
resentation. Otherwise we could not have obtained a consciousness of them through
this kind of process.
How is this a priori synthesis of apprehension supposed to restore the contact to
the synoptical elements below the line that separates receptivity from spontaneity - a
contact that guarantees the 'guidance from without' required by the conceptual
order? Since the qualitative aspects of our sensory consciousness above and below
the line cannot be used for this purpose, we should now heed its formal aspects not
only above the line but also below it. I have noted before that it is not possible to
give any direct characterization of these forms. But now we can substantiate the
previously empty, indirect characterization of those forms of receptivity as being of
a kind that produces synoptically structured representations, which are able to serve
as the basis for a synthesis producing exactly our forms of intuition. In other words,
we would not have synthesized exactly these forms of intuition by a synthesis that
is both a priori and transcendental unless we had been given material synoptically
formed like this "manifold which sensibility presents in its original receptivity"
(A100; my emphasis).
Though the inaccessibility of the below the line aspects is guaranteed in this
description of the relation of formal aspects above and below that line which sepa-
rates sheer receptivity from spontaneity, the contact between the two pairs of forms
(forms of receptivity and forms of intuition) is of a kind completely different from
the relation between the receptive and spontaneous sensations. This is mainly due to
the fact that the transcendental status of forms of intuition sharply distinguishes
them from objective sensations. There is a transcendental argument for the existence
of sensations (as matter of experiences), but there is no such argument as to their
concrete nature. Yet, in the case of forms of intuition, there is, in Kant's opinion,
such an argument (e.g. in the case of space, from the necessity of thinking some-
thing distinct from us).53 It is therefore exactly those forms of intuition that we have
to synthesize in order to be able to intuitively refer to objects of experience. And this
fact is explained by our sensibility being characterized by exactly these forms of
receptivity.
Those representations then, which we de facto entertain in perception of empirical
reality, are representations dependent on the synoptically given material that is
stmctured by the forms of receptivity in their individual forma! aspects - their indi-
vidual shape and their individual duration. Furthermore, what is structured in this
process is in turn due to our being affected by things-in-themselves. Given both
these claims, it seems reasonable to conclude that the idea of a conceptually, and
hence normatively, laden process being 'guided from without' is done justice in the
Kantian system.
Cf. Haag (2007, 142-150).
5 Kant on Imagination and the Natural Sources of the Conceptual 85
References
Haag, Johannes. 2007. Erfahrung und Gegenstand. Das Verhältnis von Sinnlichkeit und Verstand.
Frankfurt a.M: Vittorio Klostermann.
Henrich, Dieter. 1976. Identität und Objektivität. Eine Untersuchung über Kants transzendentale
Deduktion. Heidelberg: Winter.
Kant's works are quoted in the usual manner after Kants gesammelte Schriften, vols. I-XX1I ed.
Preußischen Akademie der Wissenschaften, Berlin 1902 ff., vol. XXIII ed. Deutschen Akademie
der Wissenschaften, Berlin, 1956, vols. XXIV-XXIX ed. Akademie der Wissenschaften zu
Göttingen, Berlin 1966 ff. The translation mostly follows the Cambridge edition ofthe works
of Immanuel Kant, eds. F. Guyer and Allen W, Wood.
McDowell, John. 2009. Sensory consciousness in Kant and Sellars. In Having the world in view.
Essays on Kant. Hegel, and Sellars, ed. John McDowell, 108-126. Cambridge, MA: Harvard
University Press.
Rosenberg, Jay. 2007a. Divergent intuitions: McDowell's Kant and Sellars's Kant. In Wilfrid
Sellars: Fusing the images, ed. Jay Rosenberg, 266-290. Oxford: Oxford University Press.
Rosenberg, Jay. 2007b. Perception vs. inner sense: A problem about direct awareness. In Wilfrid
Sellars: Fusing the images, ed. Jay Rosenberg, 230-242. Oxford: Oxford University Press.
Sellars, Wilfrid. 1963. Phenomenalism. In Science, perception and reality, ed. Wilfrid Sellars,
60-105. London: Routledge and Kegan Paul.
Sellars, Wilfrid. 1968. Science and metaphysics. Variations on Kantian themes. London: Routledge
& K. Paul.
Sellars, Wilfrid. 1976. Kant's transcendental idealism. Co//ecrio«5 o/PW/oi-op/ry 6: 165-181.
Sellars, Wilfrid. 1978a. The role of imagination in Kant's theory of experience. In Categories: A
colloquium, ed. H. Johnstone, 231-245. University Park: Pennsylvania State University.
Sellars, Wilfrid. 1978b. Berkeley and Descartes: Reflections on the theory of ideas. In Studies in
perception. Interrelations in the history of philosophy and science, ed. P. Machamer and R.
Turnbull, 260-311. Columbus: Ohio State University Press.
Sellars, Wilfrid. 1982. Sensa or sensings: Reflections on the ontology of perception. Philosophical
Studies A\:%-i-\\\.
Vaihinger, Hans. 1892. Commentar zu Kants Kritik der reinen Vernunft. Stuttgart (Bd. 1: 1881; Bd.
2:1892).
Waxman, Wayne. 1991. Kant's model ofthe mind. Oxford: Oxford University Press.
Wolff, Robert P. 1963. Kant 's theory of mental activity: A commentary on the transcendental analytic
ofthe 'critique of pure reason'. Cambridge, MA: Harvard University Press.